Three new species of spiny throated reed frogs(Anura: Hyperoliidae) from evergreen forestsof TanzaniaLoader et al.

Loader et al. BMC Research Notes (2015) 8:167 DOI 10.1186/s13104-015-1050-y

Loader et al. BMC Research Notes (2015) 8:167 DOI 10.1186/s13104-015-1050-y

RESEARCH ARTICLE Open Access

Three new species of spiny throated reed frogs(Anura: Hyperoliidae) from evergreen forestsof TanzaniaSimon P Loader1*, Lucinda P Lawson2,3,4, Daniel M Portik5 and Michele Menegon6

Abstract

Background: The East African spiny-throated reed frog complex (Hyperolius spinigularis, H. tanneri, and H. minutissimus)is comprised of morphologically similar species with highly fragmented populations across the Eastern AfromontaneRegion. Recent genetic evidence has supported the distinctiveness of populations suggesting a number of crypticspecies. We analyse newly collected morphological data and evaluate the taxonomic distinctiveness of populations.

Results: We find three new distinct species on the basis of morphological and molecular evidence. The primarymorphological traits distinguishing species within the Hyperolius spinigularis complex include the proportions anddegree of spinosity of the gular flap in males and snout-urostyle length in females. Other features allow the threespecies to be distinguished from each other (genetics). We refine the understanding of H. minutissimus which can befound in both forest and grassland habitats of the Udzungwa Mountains, and provide more details on the call of thisspecies. Further details on ecology are noted for all species where known.

Conclusions: Three new species are described and we narrow the definition and distribution of Hyperolius spinigularisand H. minutissimus in East Africa. The spiny-throated reed frogs have highly restricted distributions across the fragmentedmountains of the Eastern Afromontane region. Given the newly defined and substantially narrower distributions of thesespiny-throated reed frog species, conservation concerns are outlined.

BackgroundThe East African spiny-throated reed frog complex(Hyperolius spinigularis, H. tanneri, and H. minutissi-mus) is comprised of morphologically similar speciesoccupying isolated mountaintops across the EasternAfromontane region. Lawson [1] provided molecular evi-dence supporting the recognition of these three taxa andtheir distinctiveness from one another. The validity ofthese species has never been seriously questioned, thoughsuggestions of more than one species in Hyperolius spini-gularis and H. minutissimus have been remarked upon inthe literature [1,2].Hyperolius spinigularis was described by Stevens [3]

based on material collected from Mulanje in Malawi.Subsequently, Schiøtz [2] reported the presence of thisspecies ca. 1300 km north in East Usambara Mountains

* Correspondence: simon.loader@unibas.ch1Department of Environmental Sciences, University of Basel, BiogeographyResearch Group, Basel 4056, SwitzerlandFull list of author information is available at the end of the article

© 2015 Loader et al.; licensee BioMed Central.Commons Attribution License (http://creativecreproduction in any medium, provided the orDedication waiver (http://creativecommons.orunless otherwise stated.

in Tanzania, though questioned its taxonomic place-ment. Schiøtz [2] noted (p. 166) “it is questionablewhether the two forms should be separated subspecifi-cally.” Schiøtz expanded on morphological differencesbetween the populations stating “the males are of thesame size, the females seem smaller in the northernsample. The breadth of the protective flap is greater thanthe length and often weakly bilobed in the type material,circular or almost circular in the northern sample” [2],p. 166). This initial observation of geographically distinctmorphological variation cast doubt on whether theMalawi and northern Tanzanian populations were part ofthe same lineage or better regarded as distinct species.Since its initial description, Hyperolius spinigularis has

been documented to occur across a much larger range inEast Africa beyond Malawi and East Usambara. This in-cludes Udzungwa [4], Nguru [5] and Uluguru [1]. How-ever, the record provided by Schiøtz and Westergaard [4]of H. spinigularis in the forests of the Udzungwa was in-correct. The Udzungwa species record is Hyperolius

This is an Open Access article distributed under the terms of the Creativeommons.org/licenses/by/4.0), which permits unrestricted use, distribution, andiginal work is properly credited. The Creative Commons Public Domaing/publicdomain/zero/1.0/) applies to the data made available in this article,

Loader et al. BMC Research Notes (2015) 8:167 Page 2 of 16

minutissimus – a species Schiøtz and Westergaard [4] de-scribed. H. minutissimus is found across forest and grass-land sites in the Udzungwa. Tellingly, Schiøtz andWestergaard [4] (p. 8) noted about their Udzungwa forestmaterial: “A chirping voice, acoustically similar to that ofH. minutissimus, was noted from specimens kept in plasticbags.” Also noting on their collection (p. 8), “The new ma-terial seems in these characters closest to the southernpopulation; the gular flap is slightly heart-shaped withwidth greater than length in most specimens” which isalso in line with the broader proportions of the gular flapin H. minutissimus. Beyond the Udzungwas, Portik et al.[6] reported H. cf. spinigularis from the Namuli massif inMozambique, and Lawson [1] detailed the occurrence ofH. spinigularis from the Uluguru in Tanzania. Overall therange of species referable to H. spinigularis has been re-corded from fragmented and distant locations across theelements of the Afromontane region of East Africa. Thespecies H. minutissimus was described by Schiøtz [2]10 km West from Njombe in the vicinity of the SouthernHighlands with populations also recorded further northfrom the grasslands [4] and forests [1] of the Udzungwa.Lawson [1] outlined considerable genetic diversity in the

species H. minutissimus and H. spinigularis above the spe-cies level. This included samples from recent surveys inTanzania in the Rubeho and Livingstone Mountains andin Mozambique (Namuli) by the authors of this paper.Analyses by Lawson et al. [7] provided further evidence ofthese divergences, which sampled across known popula-tions and extended the geographic scope of these analyses.These recently discovered genetically divergent line-

ages also prove to be diagnosable using morphologicalcharacters, and here we provide formal species descrip-tions and a revised identification key for these lineages.We reassess the geographic distributions of the spiny-throated reed frogs based on these discoveries – clarify-ing previous uncertainty in the distribution of species,and address the conservation implications based on thenewly defined and narrowed ranges.

MethodsMolecular data and analysisSpecimens collected from our fieldwork were fixed in ei-ther 95% ethanol or 5% formalin, and subsequentlystored in 70% ethanol. Samples of muscle and/or liverwere taken from representative individuals and pre-served in 95% ethanol, these specimens are listed inAdditional file 1. Lawson [1] and Lawson et al. [7] pro-vide details on the approaches and genes used in thisstudy, which include one mitochondrial gene and threenuclear loci (mitochondrial: NADH dehydrogenase sub-unit 2; nuclear: POMC, C-myc and Rag-1). In Lawson etal’s [7] publication their Table 1 provide details on sam-ples included in this study, including their origin and

associated Genbank numbers. Phylogenetic relationshipswere estimated between all individuals using likelihoodand Bayesian methods, including BEAST, RAxML, andBPP [8-10], using data from Lawson et al. [7]. Speciestrees were constructed based on these trees and by usingspecies delimitation methods in *BEAST and BPP [10,11].To examine species boundaries across the reconstructedphylogeny we applied three species delimitation methods:a Bayesian implementation of the General Mixed Yule-Coalescent model (“bGMYC” package v. 1.0.2 for R,[12]) using trees from the BEAST analysis, a Bayes Fac-tor species Delimitation (BFD;[13]) to compare alter-native scenarios for the H. minutissimus, H.spingularis, and H. tanneri species complex using al-ternate *BEAST species trees (See Additional file 1:Table S1), and a joint estimation of the species tree andspecies delimitation in BPP3 [10]. Each of the coales-cent species tree analyses were run twice.

MorphologyMaterial was examined in the following institutions: TheNatural History Museum, London (BMNH); FieldMuseum, Chicago (FMNH); California Academy of Sci-ences, San Francisco (CAS); Museo Tridentino di ScienzeNaturali, Trento (MTSN), and University of Dar es Salaam(UDSM) (see in Additional file 1: Table S1). Morphologicalmeasurements were taken using dial callipers, to the near-est 0.1 mm using Mitutoyo Absolute Digimatic Calipers(CD-6”C) with the aid of a Leica MZ8 stereo microscope(Leica Microsystems GmbH, Wetzlar, Germany). Onlyfully-grown specimens (adult color pattern and adult size)were measured. Sex was determined by the presence orabsence of gular flap in adult specimens. Measurements inthis analysis were length: Snout-Urostyle Length (SUL),Head Width (HW), Head Length Diagonal from corner ofmouth (HLD), Head Length Diagonal from jawbone end(HLDJ), Nostril-Snout (NS), Inter-narial (IN), Eye toNostril (EN), Eye Distance (EE), Inter-orbital (IO), Tibiafi-bula Length (TL), Thigh Length (THL), Tibiale FibulareLength (TFL), Foot Length (FL), Forelimb Length (FLL),Hand Length (HL), Width of Gular Flap (WGF) Height ofGular Flap (HGF). Furthermore, qualitative characterswere investigated: gular shape, proportions, and spinosityto assess species differences. In order to assess the mor-phometric distinctness of these species we also conductedPrincipal Component analyses on log-transformed datausing various packages in R [14-16]. See Lawson et al. [7]for further explanation of these results.

Acoustic informationAdvertisement calls were recorded with an Olympus LS-10PCM digital stereo audio recorder equipped with a Sonydirectional microphone. The calls were analyzed using thesoftware package Raven 1.2 [17].

Table 1 Descriptive statistics for the three new species for 17 morphological characters

Males

Species SUL HW HLD HLDDJ NS IN EN EE IO TL THL TFL FL FLL HL WGF HGF

H. burgessi (19)

Average 18.5 6.4 5.8 6.5 1.0 1.9 1.9 3.8 2.4 9.2 8.8 5.9 7.7 4.4 5.0 5.0 5.1

St. Dev 1.3 0.4 0.4 0.5 0.1 0.1 0.1 0.5 0.2 0.5 0.5 0.3 0.5 0.4 0.2 0.4 0.4

Range 16.4-20.3 5.4-7.3 5.2-6.5 5.9-7.4 0.9-1.2 1.6-2.1 1.7-2.3 2.6-5 2.2-3 8.4-9.8 8-9.9 5.1-6.3 6.7-8.6 3.9-5.8 4.5-5.5 4.2-6.1 4.3-6.1

H. davenporti (10)

Average 18.9 6.5 5.4 6.4 1.0 1.9 1.9 3.8 2.4 9.1 9.1 5.8 7.4 4.4 5.4 5.4 5.0

St. Dev 1.0 0.3 0.3 0.4 0.1 0.1 0.1 0.2 0.2 0.4 0.4 0.4 0.7 0.2 0.3 0.3 0.4

Range 17.3-20.2 6-7.1 5-6.1 5.6-6.9 0.9-1.2 1.8-2.1 1.8-2 3.4-4 2.1-2.7 8.4-9.8 8.4-9.7 5-6.4 6.2-8.5 4.1-4.7 5.1-5.8 4.6-5.8 4.4-5.6

H. minutissimus (13)

Average 20.3 6.8 6.1 6.5 1.0 2.0 2.0 4.2 2.4 9.9 9.5 6.2 8.8 4.7 5.8 5.9 4.5

St. Dev 1.2 0.5 0.4 0.0 0.1 0.1 0.1 0.4 0.2 0.6 0.6 0.5 0.7 0.4 0.4 0.4 0.4

Range 18.8-22.7 6-7.5 5.5-6.8 6.5-6.5 0.9-1.1 1.7-2.1 1.7-2.3 3.4-4.7 1.9-2.8 8.7-11 8.6-10.7 5.5-7.3 7.8-10.3 4-5.5 5-6.8 5.2-6.9 3.7-5.1

H. spinigularis (9)

Average 19.6 6.9 6.1 6.9 1.1 1.9 2.0 3.9 2.5 9.9 8.8 6.2 8.2 4.5 5.3 5.1 4.0

St. Dev 1.0 0.4 0.4 0.4 0.1 0.1 0.2 0.2 0.3 0.3 0.6 0.4 0.7 0.2 0.4 0.4 0.3

Range 18.6-21.8 6.2-7.6 5.8-7.1 6.5-7.8 1-1.2 1.7-2.1 1.7-2.2 3.7-4.3 2-2.9 9.3-10.2 7.6-9.6 5.6-6.9 7.6-9.6 4.2-5 4.7-5.9 4.3-5.5 3.6-4.4

H. tanneri (2)

Average 22.2 6.9 6.1 6.9 1.2 2.0 2.1 4.1 2.7 11.1 8.9 6.0 9.5 4.7 6.3 NA NA

St. Dev 0.4 0.9 0.6 0.5 0.1 0.2 0.3 0.8 0.4 0.7 1.6 1.1 1.1 0.4 0.7 NA NA

Range 21.9-22.5 6.2-7.5 5.7-6.5 6.5-7.2 1.2-1.1 1.8-2.1 2.3-1.9 3.5-4.7 2.4-3 11.6-10.6 7.7-10 5.2-6.7 8.7-10.3 4.4-5 5.8-6.8 NA NA

Females

H. burgessi (28)

Average 24.3 8.5 7.2 8.3 1.3 2.4 2.3 4.8 3.1 12.1 11.2 7.6 10.3 5.4 7.1

St. Dev 1.2 0.5 0.6 0.7 0.1 0.1 0.3 0.3 0.3 0.8 0.9 0.4 0.7 0.3 0.4

Range 21.3-25.9 7.6-9.6 6.1-8.3 6.5-9.4 1-1.5 2.1-2.6 1.7-2.8 4-5.3 2.3-3.8 10.6-13.1 9.8-12.9 6.9-8.4 9-11.7 4.6-5.9 6-7.8

H. davenporti (2)

Average 27.0 9.2 6.9 8.4 1.3 2.3 2.6 5.2 2.8 12.9 13.0 8.6 10.7 6.3 7.4

St. Dev 0.1 0.0 0.0 0.3 0.1 0.1 0.1 0.2 0.1 0.4 0.6 0.1 0.3 0.7 0.2

Range 26.9-27 9.2-9.2 6.9-6.9 8.2-8.6 1.2-1.3 2.2-2.3 2.5-2.7 5-5.3 2.7-2.9 12.6-13.1 12.5-13.4 8.5-8.6 10.5-10.9 5.8-6.8 7.2-7.5

Loaderet

al.BMCResearch

Notes

(2015) 8:167 Page

3of

16

Table 1 Descriptive statistics for the three new species for 17 morphological characters (Continued)

H. ukwiva (2)

Average 28.7 10.0 8.0 9.8 1.5 2.8 2.7 5.5 3.2 14.2 14.1 8.9 12.7 7.5 9.0

St. Dev 0.9 0.8 0.1 0.7 0.2 0.4 0.1 0.2 0.1 1.1 0.8 0.2 0.5 0.4 0.8

Range 28-29.3 9.4-10.5 7.9-8.1 9.3-10.3 1.3-1.6 2.5-3 2.6-2.8 5.3-5.6 3.1-3.3 13.4-15 13.5-14.6 8.7-9 12.3-13 7.2-7.7 8.4-9.6

H. spinigularis (3)

Average 24.5 8.7 7.3 8.8 1.3 2.3 2.6 4.9 3.1 12.9 11.8 8.2 11.1 5.6 7.0

St. Dev 1.1 0.6 0.5 0.6 0.0 0.1 0.1 0.1 0.3 0.3 0.6 0.5 0.8 0.4 0.4

Range 23.6-25.7 8.3-9.4 6.9-7.8 8.3-9.5 1.3-1.3 2.3-2.4 2.5-2.7 4.8-5 2.8-3.4 12.7-13.3 11.4-12.5 7.7-8.6 10.2-11.8 5.3-6 6.6-7.3

H. tanneri (2)

Average 26.7 9.2 7.6 8.7 1.5 2.4 2.7 5.2 2.9 15.0 11.2 7.7 10.7 6.1 7.4

St. Dev 4.3 1.5 1.1 0.7 0.3 0.5 0.2 0.5 0.4 5.7 3.3 2.3 3.0 1.7 1.5

Range 23.6-29.7 8.1-10.2 6.8-8.4 8.2-9.2 1.3-1.7 2-2.7 2.5-2.8 4.8-5.5 2.6-3.2 10.9-19 8.8-13.5 6.1-9.3 8.5-12.8 4.9-7.3 6.3-8.4

Loaderet

al.BMCResearch

Notes

(2015) 8:167 Page

4of

16

Loader et al. BMC Research Notes (2015) 8:167 Page 5 of 16

Results and DiscussionPhylogeny and species delimitationAll methods agreed on the optimal resolution of theevolutionary relationships within the clade of spiny-throated frogs, which consists of a well-supportedmonophyletic assemblage (Figure 1). Hyperolius minutis-simus and H. ukwiva sp. nov. form a group which is asister group to H. davenporti sp. nov., H. burgessi sp.nov., H. tanneri, and H. spinigularis. Within the latterclade it is shown that H. tanneri is well supported as thesister species to H. davenporti sp. nov., H. burgessi sp.nov., and H. spinigularis. Hyperolius davenporti sp. nov.and H. burgessi sp. nov. form a close grouping. Speciesdelimitation approaches agreed with the taxonomic unitsrecognized and clades recovered in the phylogeny (seeSystematics section below: Additional file 1: Table S1)(see also Lawson, et al. [7].

MorphologyMeasurements for specimens are given in Additional file2. Summary statistics of each species and characters aregiven in Table 1. Many traits distinguish species includ-ing SUL (M/F), TL/SUL (M/F), HW/SUL (M), GFH/SUL (M), GFW/GFH (F) (Figure 2, Table 2). PCA ana-lysis of males shows largely overlapping results, thoughH. burgessi sp. nov. and H. minutissimus are distinct (see

Figure 1 Bayesian maximum clade credibility chronogram of the spiny-thrbranches as well as location of clades.

Lawson et al. [7]). PC1 is evenly representative of alltraits. PCA analysis of females shows large areas of over-lap, though the two H. ukwiva sp. nov. and two H. tan-neri individuals are outside of the centroids of overlapfor other species. PC1 represents 62% of variance and iseven across traits with strongest influence from headwidth, snout-urostyle length, and leg measurements(see Supplementary Data Figure S3 and Table three inLawson et al. [7]). Discriminant function analysis ofmales showed complete segregation between species,and females were distinguished for all species exceptfor H. spinigularis and H. burgessi sp. nov., which wereentirely overlapping (Supplementary Table S3, Lawsonet al. [7]). Differentiation in males is largely driven byheight of the gular flap (HGF). In females, differenti-ation is strongly dominated by tibia length and snout-urostyle length.

SystematicsHyperolius burgessi sp. nov.Holotype BMNH 1974.299 (male) collected in AmaniNature Reserve, East Usambara Mountains, Tanzania(−5.2 S, 38.6167 E) by Alice Grandison (Figure 3).

Paratypes We restrict paratype material to localitieswithin the East Usambara on the basis that further

oated reed frog species complex. Support for clades is shown on

Figure 2 Plots of Snout – Urostyle Length (SUL), TL/SUL, HW/SUL, GFW/SUL, GFH/SUL, and GFW/GFH. Plots are grouped according to speciesand sex where appropriate, and color-coded following Figures 1 and 5. Boxes represent the interquartiles. Whiskers extend to the most extremedata point no more than 1.5 times the interquartile range from the box. Width of each box reflects relative sample sizes (square-roots of thenumber of observations). Asterix mark the holotype measurement.

Loader et al. BMC Research Notes (2015) 8:167 Page 6 of 16

detailed morphological/molecular analysis might un-cover additional cryptic lineages. Males: BMNH1974.295-298, BMNH 1974.300-303, BMNH 1974.304,same collection data as holotype; CAS 169977, CAS169979–169980, CAS 169982–169985, CAS 169987–169990, CAS 169992–169994, CAS 169997–169998,CAS 170000–170005 collected in East Usambara Moun-tains, Tanzania by Robert Drewes and Jens Vindum,April 1988; FM 274310–274312, FM 274322 collected inEast Usambara Mountains, Tanzania by Lucinda Lawson.Females: CAS 169258–169262, CAS 169945–169946,CAS 169976, CAS 169981, CAS 169991, CAS 169995–169996 collected in various localities in East UsambaraMountains, Tanzania by Robert Drewes and JensVindum, April 1988; FM 274313–274314, FM 274320–274321, FM 274323–274324 collected in in various lo-calities in East Usambara Mountains, Tanzania byLucinda Lawson.

Referred material Males: MTSN 8238, MTSN 8240–8241, MTSN 8247, MTSN 8259–8260, MTSN 8266–8267 collected in Pemba, Nguru Mountains, Tanzania byMichele Menegon; FM 274259–60 collected in UluguruNature Reserve, Uluguru North, Uluguru Mountains,Tanzania by Lucinda Lawson. Females: MTSN 8265,MTSN 8271, MTSN 8273, MTSN 8278 collected inNguru Mountains, Tanzania by Michele Menegon; FM274258 collected in Uluguru Nature Reserve, UluguruNorth, Uluguru Mountains, Tanzania by LucindaLawson.

Diagnosis Horizontal pupil with distinctive gular flap inmales. As with most other members of the clade of spiny-throated frogs (Hyperolius spinigularis, H. davenporti, H.minutissimus), H. burgessi also has the presence of dermal

asperities (including the body and chin region) on the ven-trum, unique amongst hyperoliids. The presence of asper-ites on the gular flap diagnoses this species from H.tanneri, for which they are absent. The even distributionof dermal asperites on the gular flap differs from the an-teriorly positioned distribution of asperites in H. minutis-simus and H. ukviwa. Furthermore, in males, the specieshas a distinctive gular flap morphology which differs fromother members of the genus. H. burgessi males have arounded gular flap which is not bilobed - distinctive fromH. spinigularis. The shape of the gular flap also narrowsanteriorly, being much smaller in size to the base of thegular flap, which is different from the gular flap of H.davenporti that is more equal in size anteriorly and pos-teriorly. Furthermore, the shape of the gular flap in malesis different than H. davenporti in usually having an equalor greater height than width (see Figures 2, 3 and 4;Tables 1 and 2). Based on molecular comparisons the spe-cies is genetically distinct from close relatives (H. burgessi,H. spinigularus, and H. tanneri,), and is minimally 2%pairwise divergent from its closest relative based onmtDNA (ND2. Table 3; see Figure 1). Hyperolius burgessialso has a largely allopatric distribution with respect toother species in the complex (Figure 5).

Description of holotype Small to moderate sized hyper-oliid. Pupil horizontal. Snout blunt slightly rounded.Canthus rostralis angular, being slightly convex on thehorizontal plane and slightly concave on the verticalplane. Distance between eyes is 4.1 mm and the inter or-bital distance is 2.4 mm. The inter-narial distance is2.0 mm, greater than narial distance to the eye(1.8 mm). The nostril to snout is 1.0 mm. The width ofhead (6.9 mm) equaling 0.36 length of body (19.2 mm).The gular flap width is less than (5.0 mm) the height

Table 2 AMOVA tables for each morphometric variable/ratio for species with sex as a covariate

Females

SUL Df Sum Sq Mean Sq F value Pr (>F)

Species 4 0.079 0.020 5.973 0.001

Residuals 32 0.105 0.003

TL/SUL Df Sum Sq Mean Sq F value Pr (>F)

Species 4 0.023 0.006 0.323 0.861

Residuals 32 0.570 0.018

HW/SUL Df Sum Sq Mean Sq F value Pr (>F)

Species 4 0.002 0.001 0.286 0.885

Residuals 32 0.065 0.002

Males

SUL Df Sum Sq Mean Sq F value Pr (>F)

Species 4 0.116 0.029 7.941 0.000

Residuals 48 0.176 0.004

TL/SUL Df Sum Sq Mean Sq F value Pr (>F)

Species 4 0.013 0.003 1.641 0.179

Residuals 48 0.094 0.002

HW/SUL Df Sum Sq Mean Sq F value Pr (>F)

Species 4 0.037 0.009 1.962 0.115

Residuals 48 0.224 0.005

GFW/SUL Df Sum Sq Mean Sq F value Pr (>F)

Species 4 0.126 0.032 5.736 0.001

Residuals 47 0.258 0.005

GFH/SUL Df Sum Sq Mean Sq F value Pr (>F)

Species 4 0.816 0.204 51.880 0.000

Residuals 47 0.185 0.004

GFW/GFH Df Sum Sq Mean Sq F value Pr (>F)

Species 4 0.905 0.226 84.290 0.000

Residuals 47 0.126 0.003

Df = Degrees of Freedom. Sq = squares. F value = F statistic. Pr (>F) = probabilityof F test (p value).

Loader et al. BMC Research Notes (2015) 8:167 Page 7 of 16

(5.1 mm). The gular flap is rounded, thickened and notbilobed, anteriorly narrowing so that overall shape iswide based hexagon. It is marked by black asperities (ca.80) evenly distributed across the whole of the gular flap.Some asperities, sparsely distributed, though more con-centrated, on the apex of the chin (mentum). Tibio-tarsal articulation of the adpressed hind limb reachingthe eye. Tibio-tarsal (9.1 mm) is almost equal to thighlength (8.5 mm). The tibiable fibulare length is 5.9 mm.The toes have expanded fleshy discs with the foot being7.8 mm. Webbing is extensive reaching the base of thefleshy discs on all toes apart from the first toe where itonly reaches the first tubercle. The forelimb length is4.2 mm, less than the hand length (5.0 mm). The handshave expanded, rounded fleshy discs. Webbing justreaching distal subarticular tubercle of the outer finger,

reaching distal subarticular tubercle of the 4th toe onboth sides. Dorsal skin surface granular with a single mi-nute black asperites surmounting many of the granules.Ventral skin surface strongly granular with black asper-ities restricted to the mentum, gular disc, abdomen andundersurfaces of the femur. Ventral asperities muchmore prominent than those of the dorsum.

Paratypes Head and body proportions in close agree-ment with those of the holotype (Figure 2, Table 1). Thedistribution of the asperities of the males are in closeagreement with that of the holotype. The proportions ofthe gular flap, diagnostic for the species, shows somevariation which means care needs to be taken in apply-ing this character (Figure 2). Webbing of all the materialconforms to that of the holotype. Material from theUluguru mountains (FM 274258–60) is strongly dehy-drated and this might have impacted the morphometricmeasurements. Uluguru material show extreme valuesfor gular flap proportions. Freshly collected material willbe necessary to assess the morphometric variationamong populations that might potentially recognize oneof these populations as being distinct. Given the largeamount of molecular difference in Uluguru populations(1.7% mtDNA (ND2) pairwise divergence from joint EastUsambara and Nguru populations) the population maybe a candidate for a new species.

Colour patterning of adults in life See Figure 6 forphoto in life. Generally, the females and males resemblethe holotype in basic coloration – not showing dichro-matic patterns from samples collected. The brown dorsalchromatophores varied in intensity from specimen, andvaried within and between population (Nguru and EastUsambara for which large series exist) (see Additionalfile 1: Figure S1 and S2). It should be noted that theNguru material was prepared differently and much morerecently than most of the East Usambara material com-pared. Some differences might reflect preservation differ-ences. The intensity of chromotophores sometimesresulted in dark brown mottling (particularly in EastUsambara material – see Additional file 1: Figure S1).The majority of specimens either had lateral dark edgedwhite stripes (either thin or irregular in size and outline)ending anteriorly in a narrow stripe meeting at the snoutor a triangle covering the snout. The ventral side is of alighter cream coloration.

Sexual dimorphism Females attaining a much largersize than the males (Figure 2). Asperities of the dorsumare less visible in the female and absent from the ventralside in females. Males are easily distinguished from the fe-males during the breeding season by their characteristicrounded, slightly narrowing anteriorly gular sac (Figure 4).

Figure 3 Dorsal and ventral views of the holotype of H. burgessi BMNH 1974.299. Bar = 5 mm.

Loader et al. BMC Research Notes (2015) 8:167 Page 8 of 16

Advertisement call No advertisement call is known butStevens [3] reported the males making a “weak, raspinghigh-pitched “tcheek-tcheek” call, believed to serve a ter-ritorial function in the close relative H. spinigularis oc-curring in Malawi and Mozambique. Stevens [3]questioned whether this might also be an advertisementcall but speculated that the restricted breeding area andseason in this species might have “obviated the necessityof a mating call.” Vonesh (in litt.) conducted intensivesurvey of this species over a two-year period in AmaniNature Reserve and never heard calling males, which issuggestive of a lack of advertisement or territorial callsin H. burgessi.

Etymology The species is named for Prof. Neil Burgess,who has made and continues to make enormous effortstowards conserving Tanzania’s forests, which this newspecies survival depends upon. The species is also re-stricted to the Eastern Arc Mountain region – an areawhich Neil has particularly devoted considerable timeand energy to understand and preserve.

Distribution and conservation The species is known tooccur in East Usambara, Nguru and Uluguru Mountainsin Tanzania (Figure 5, Table 4). The species has beencollected at high altitudes in, and on the edges of sub-montane forests. Vonesh [18] (p. 281) commented onthe ecology and behavior of this species (then referred toas H. spinigularis) saying “it breeds during both annualrainy seasons by attaching its eggs to vegetation over-hanging permanent or semi-permanent ponds or

swamps. Mean clutch size is 89 eggs. During the longrains of 2002, over 75% of reproductive activity occurredduring the first 30 days of the rainy season early Marchto early April.” Observations of females attaching eggs tovegetation overhanging permanent or semi-permanentponds or swamps were also made in the Nguru popula-tions of H. burgessi (MM, pers. obs.).

Hyperolius davenporti sp. nov.Holotype MTSN 7465 (male) collected in SakaraNyumo Forest Reserve, Livingstone Mountains,Southern Highlands, Tanzania (−9.8389 S, 34.60781 E,2010 m) on 14th January 2011 by Michele Menegon(Figure 7).

Paratypes Females: MTSN 7453, and MTSN 7464.Males: MTSN 7455, MTSN 7456, MTSN 7457–7463,and MTSN 7467. Juvenile: MTSN 7466. Same collectiondata as holotype.

Diagnosis Horizontal pupil with distinctive gular flap inmales. As with most other members of the spiny-throated clade (H. spinigularis, H. burgessi, H. minutissi-mus,), H. davenporti also has the presence of dermalasperities (including the body and chin region) on theventrum, unique amongst Hyperolius. The presence ofasperites on the gular flap diagnoses this species fromH. tanneri, for which they are absent. The even distribu-tion of dermal asperities across the gular flap differsfrom the anteriorly and medially distributed asperities in

Figure 4 Schematic drawings of the ventral view of head region of spiny throated complex group.

Loader et al. BMC Research Notes (2015) 8:167 Page 9 of 16

both H. minutissimus and H. ukviwa. Furthermore, inmales, the species has a distinctive gular flap different invarious combinations to other members of the genus.H. davenporti males have a rounded gular flap which isnot bilobed, distinguishing it from H. spinigularis(Figures 2 and 4). Furthermore, the shape of the gularflap in males is different from H. burgessi in being widerthan the height and shaped more equally in the anteriorand posterior ends of the flap (Figures 2 and 7, Tables 1and 2). Based on molecular comparisons the species isalso genetically distinct from close relatives, and is min-imally 5.7% pairwise divergent from its closest relative,based on mtDNA (ND2) (Table 3; see Figure 1). Hypero-lius davenporti has an allopatric distribution with re-spect to all other species in the complex (Figure 5).

Table 3 Average nucleotide divergences in mitochondrialdata (ND2) between species and populations

1 2 3 4 5 6

1. H. burgessi(East Usambara + Nguru)

2. H. burgessi (Uluguru) 1.7%

3. H. tanneri 6.4% 6.5%

4. H. minutissimus 9.0% 9.1% 8.3%

5. H. spinigularis 7.1% 6.6% 5.8% 7.9%

6. H. ukwiva 12.5% 11.9% 10.8% 7.0% 10.9%

7. H. davenporti 2.0% 2.3% 5.7% 9.0% 5.8% 11.3%

Description of holotype Small to moderate sized hyper-oliid. Horizontal pupil. Snout blunt slightly rounded.Canthus rostralis angular, being slightly convex on thehorizontal plane and slightly concave on the verticalplane. Distance between eyes is 4.0 mm and the inter or-bital distance is 2.4 mm. The inter-narial distance is1.9 mm, almost subequal to the narial distance to theeye (2.0 mm). The nostril to snout is 1.0 mm. The widthof head (7.1 mm) equaling 0.37 length of body(19.1 mm). The gular flap is wider (5.3 mm) by 1.10,than it is in height (4.8 mm). The gular flap is rounded,thickened, and not bilobed. It is marked by black asper-ites (ca. 50) evenly distributed across the whole of thegular flap. Some asperites, sparsely distributed, thoughmore concentrated, on the apex of the chin (mentum).Tibio-tarsal articulation of the adpressed hind limbreaching the eye. Tibio-tarsal (9.3 mm) is almost equalto thigh length (9.0 mm). The tibiable fibulare length is5.8 mm. The toes have expanded fleshy discs and thefoot is 6.2 mm. Webbing is extensive reaching the baseof the fleshy discs on all toes apart from the first toewhere it only reaches the first tubercle. The forelimblength is 4.6 mm, less than the hand length (5.5 mm).The hands have expanded, rounded fleshy discs. Web-bing just reaching distal subarticular tubercle of theouter finger, reaching distal subarticular tubercle of the4th toe on both sides. Dorsal skin surface granular witha single minute black asperities surmounting many ofthe granules. Ventral skin surface strongly granular with

West Usambara Mountains

Nguru Mountains

Uluguru MountainsRubeho

Mountains

UdzungwaMountains

UdzungwaMountains

SouthernHighlands

MulanjeMassif

MountNamuli

East Usambara Mountains

Figure 5 Elevational map and distribution of the six spiny-throated species in East Africa. Colour codes for species are: red =H. tanneri, purple =H. burgessi,green =H. minutissimus, dark blue =H. davenporti, light blue =H. spinigularis, and black=H. ukwiva. Mountain blocks containing spiny-throated speciesare referenced.

Loader et al. BMC Research Notes (2015) 8:167 Page 10 of 16

black asperities restricted to the mentum, gular disc, ab-domen and undersurfaces of the femur. Ventral asper-ities much more prominent than those of the dorsum.

Paratypes Head and body proportions are in closeagreement with those of the holotype (Figure 2, Table 1).

The distribution of the asperities of the males are inclose agreement with that of the holotype. The propor-tions of the gular flap, diagnostic for the species, showsome variation which mean the boundaries of diagnos-ing this species are in some cases slightly overlapping(Figure 2). Webbing of all the material conforms to that

Figure 6 Colour in life of: (top row) H. burgessi from Nguru (left male, right female); (second row) H. davenporti from Livingstone Mts. (left male,right female); (third row left) H. ukwiva; (third row, right) H. spingularis from Malawi; (fourth row, left) H. minutissimus from Udzungwa Mts.;(fourth row, left) and H. tanneri from West Usambara.

Loader et al. BMC Research Notes (2015) 8:167 Page 11 of 16

of the holotype. One single specimen (MTSN 7466) hadnot fully metamorphized and measured 14.3 mm(Snout-Urostyle) with a tail length of 13.8 mm.

Colour patterning of adults in life See Figure 6 forphoto in life. Generally the females and males resemblethe holotype in basic coloration. The brown dorsal chro-matophores varied slightly in intensity amongst speci-mens (see Additional file 1: Figure S3). The intensity ofchromotophores sometimes resulted in dark brown mot-tling. The majority of specimens either had lateral darkedged white stripes (either thin, broad and regular inshape or irregular in size and outline) ending anteriorlyin a narrow stripe meeting at the snout or a triangle cov-ering the snout. The ventral side is of a lighter creamcolouration.

Sexual dimorphism Females attaining a much largersize than the males (Figure 2). Asperities of the dorsumweaker in the female and absent from the ventral side infemales. Males are easily distinguished from the females

during the breeding season by their characteristicrounded and wide gular sac (Figure 4).

Advertisement call Neither advertisement or territorialcalls in H. davenporti were recorded or heard, howeversurvey time in the area (5 days) was relatively short. Lit-tle can be concluded on whether it resembles its conge-ners H. burgessi or H. spingularis as no call is knownfrom these species.

Etymology The species is named after Dr. TimDavenport, who has made substantial contributions to-wards conserving Tanzania’s forests, in particular theSouthern Highlands and Livingstone Mountains ofTanzania. The Livingstone Mountains are the onlyknown locality of this species.

Distribution and conservation The species is onlyknown from Sakara Nyumo Forest Reserve, LivingstoneMountains, Southern Highlands (Figure 5, Table 4).Specimens were collected in shallow ponds on the forest

Table 4 Species, altitudinal range, habitat and available area of occurrence

Species Altitudinal range Habitat Expected area of occurrence

H. burgessi East Usambara: Submontane forest 14,774 km2

900–1100 m

Nguru: 900–1000 m

Uluguru: 980 m

H. davenporti Livingstone: 2010 m Montane forest edge 28 km2

H. minutissimus Njombe: 2010 m Montane forest edge and grassland 14,904 km2

Udzungwa:

1680–1970 m

H. spinigularis Malawi: 690 m Submontane forest and forest edge 5,488 km2

Mozambique: 1250 m

H. tanneri West Usambara: Submontane forest and forest edge 4 km2

1310–1650 m

H. ukwiva Rubeho: 1660 m Montane forest edge 1,179 km2

Loader et al. BMC Research Notes (2015) 8:167 Page 12 of 16

edge in a large swampy area. Species were not found inthe open and were always on the forest edge. The areain the Livingstone Mountains comprises a number offragmented forest patches where the species might alsobe found. However, based on sampling carried out thiswas the only location where this species has so far beenrecorded. Further sampling will be required to establishwhether the species is truly restricted to this single forestreserve or more broadly in the Livingstone Mountainsor even across the whole Southern Highland region.

Hyperolius ukwiva sp. nov.Holotype MTSN 5064 (KMH 35846) (female) collectedin Ukwiva Forest Reserve, Rubeho Mountains, Tanzania(−7.11186 S, 36.64058 E, at 2060 m) on 6th November2006 by Frontier-Tanzania (see Figure 8).

Paratypes Male: KMH 36053 (male), Female: MTSN5085 (KMH 36056). Same collecting locality as holotypebut collected on 5th November 2006.

Figure 7 Dorsal and ventral views of the holotype of H. davenportiMTSN 7465. Bar = 5 mm.

Diagnosis Horizontal pupil with distinctive gular flap inmales. As with most other members of the spiny-throated clade (H. spinigularis, H. burgessi, H. daven-porti, H. minutissimus), H. ukwiva also has the presenceof dermal asperities (including the body and chin region)on the ventrum, unique amongst members of the genusHyperolius. The presence of asperities on the gular flapdiagnoses this species from H. tanneri, for which theyare absent. The even distribution of dermal asperities onthe gular flap differs from the anteriorly positioned dis-tribution of asperities, also found on the chin, in H. min-utissimus and H. ukviwa. Furthermore, in males, thespecies has a distinctively shaped gular flap, differentfrom H. minutissimus in being bilobed and wider thanits height (Figures 2 and 4). Females are larger inH. ukviwa, reaching sizes >25 mm, substantially largerthan females of H. minutissimus (18–24 mm) (Figures 2and 3, Tables 1 and 2). Based on molecular comparisonsthe species is also genetically distinct from close rela-tives, and is minimally 7.0% pairwise divergent from itsclosest relative, based on mtDNA (Table 3; see Figure 1).

Figure 8 Dorsal and ventral views of the holotype of H. ukwivaMTSN 5064. Bar = 5 mm.

Loader et al. BMC Research Notes (2015) 8:167 Page 13 of 16

Hyperolius ukwiva has an allopatric distribution with re-spect to all other species in the complex (Figure 5).

Description of holotype Moderate to large sized hyper-oliid. Horizontal pupil. Snout blunt slightly rounded.Canthus rostralis angular, being slightly convex on thehorizontal plane and slightly concave on the verticalplane. Distance between eyes is 5.3 mm and the inter or-bital distance is 3.1 mm. The inter-narial distance is2.5 mm, almost subequal to the narial distance to theeye (2.6 mm). The nostril to snout is 1.3 mm. The widthof head (9.4 mm) equaling 0.34 length of body(28.0 mm). Gular disc/flap is absent. Tibio-tarsal articu-lation of the adpressed hind limb reaching the eye.Tibio-tarsal (13.4 mm) is almost equal to thigh length(13.5 mm). The tibiable fibulare length is 8.7 mm. Thetoes have expanded fleshy discs. Webbing is extensivereaching the base of the fleshy discs on all toes apartfrom the first toe where it only reaches the first tubercle.The forelimb length is 7.2 mm, less than the hand length(8.4 mm). The hands have expanded, rounded fleshydiscs. Webbing just reaching distal subarticular tubercleof the outer finger, reaching distal subarticular tubercleof the 4th toe on both sides. Dorsal skin surface granularwith sparsely distributed single minute black asperitessurmounting granules. Ventral skin surface stronglygranular, particularly on the mid ventral region withlarge rounded raised surfaces. No asperities present onventral region.

Paratypes Head and body proportions in close agreementwith those of the holotype (see Figure 2; Table 1; Additionalfile 1). The distribution of the asperities of the single male(see comment below) are medially and anteriorly concen-trated on the gular flap (see Figure 4). Webbing of all thematerial conforms to that of the holotype.

Colour patterning of adults in life See Figure 6 forphoto in life. Generally the female and male resemblethe holotype in basic coloration. The dorsum is de-scribed in field notes as being “brown with two lightgrey-brown stripes from nose to the hindlegs” for allthree specimens. The ventrum is described as sunshine“yellow”. The legs and arms are similarly colored dor-sally and ventrally.

Sexual dimorphism Females attaining a much largersize than the males (Figure 2). Asperities of the dorsumweaker in the female and absent from the ventral side infemales. Males are easily distinguished from the femalesduring the breeding season by their characteristic gularsac (Figure 4).

Advertisement call No calls were detected or recordedduring collection of these three specimens, only one ofwhich was a male.

Etymology The species is named after the forest area(Ukwiva) from where the type series was collected. Thespecific epithet is considered to be a noun inapposition.

Distribution, ecology and conservation The species isonly known from Ukwiva Forest Reserve, in the RubehoMountains (Figure 5; Table 4). Specimens were collectedin and around the edge of montane forest. Collectingacross the Rubeho Mountains, although only relativelyrecent, has been quite extensive (Rovero, et al. [19] so itslocalized distribution might not just be a function ofrestricted sampling.

Comment It should be noted that we would have pre-ferred to describe the holotype as a male (e.g. KMH36053), in line with the designation of males for all othermembers of this group of hyperoliids. However, the onlyknown male paratype (KMH 36053) currently cannot belocated at UDSM or MTSN, and thus measurementswere not possible. Data on this male were taken fromphotographs and observations made in Tanzania at thetime (LL and MM).

Further remarks on Hyperolius minutissimusDistribution and conservation Commenting on thedistribution and ecology of Hyperolius spingularisSchiøtz [20] (p.180) stated “a search for it at theUdzungwas revealed it both in forest and in very openfarmland, and it may be found wherever suitable habitatsexist in the eastern Tanzanian-Malawi highlands.” Ac-cording to our records and based on the clustering ofSchiøtz’s material with our samples of H. minutissimus,we can confirm that these comments refer only to H.minutissimus, with H. spingularis not recorded from theUdzungwas. Therefore the records given by Schiøtz andWestergaard [4] refer to H. minutissimus. On p.167Schiøtz [20] reports samples collected between Kilosaand Dabaga and these refer to localities reported inSchiøtz and Westergaard [4] and are limited toUdzungwa area – and not beyond, towards Kilosa, thatmight suggest potential presence in Rubeho of H. minu-tissimus. Currently, we only record the presence of H.ukwiva from Rubeho mountain region. H. minutissimus isfound in forest and grassland habitats in the UdzungwaMountains and Njombe, the latter in the region of theSouthern Highlands. A wider distribution of H. minu-tissimus in the Southern Highlands and beyond intoMalawi is as yet unconfirmed and future sampling willbe required to establish if it occurs in these areas.

Loader et al. BMC Research Notes (2015) 8:167 Page 14 of 16

Advertisement call The advertisement call is, as de-scribed by Schiøtz, [20], a fast series of quiet, unmelodicclicks, typically repeated about 3 times per second witha dominant frequency of 3.40 and a standard deviationof 0.02 kHz (Figure 9). Calls were recorded both indense forests and open wetlands. It is interesting tospeculate on the potential evolutionary scenarios for dif-ferences between spiny-throated species calls. The pres-ence of a call in H. minutissimus and apparent absencein congeners (H. spinigularis, H. burgessi, and possiblyH. davenporti) could be linked to habitat differences –important in determining call structure in organisms. H.minutissimus occurs in both forest and grassland habi-tats – with a relatively wide distribution – different com-pared to the forest restricted and highly localized speciesH. spinigularis, H. burgessi, and H. davenporti. It wouldbe interesting to further investigate this pattern andStevens’ [3] original speculations that absence of a mat-ing call was linked to such factors (e.g. restricted breed-ing area and season).

Key to species of the East African spiny throat reed frogsHere we present a key that should identify adult malespecimens of all presently described species. Due to theirsimilarity, identification of females through a dichotom-ous key is currently not recommended. Geographicaldistribution (and/or morphometric/molecular analysis)can help in distinguishing between the morphologicallysimilar female species. Caution is required when identi-fying small or poorly preserved specimens for which thegular flap might be damaged or desiccated:

0.6sec

kHz

kU

sec 0.6

4-

8-

0-

Figure 9 Sonogram of Hyperolius minutissimus.

1a Gular flap with black dotted asperities, species notfound in West Usambara Mountains.

21b Gular flap lacking any asperities, species found inWest Usambara Mountains.

H. tanneri2a Black dotted asperities evenly distributed across thegular flap.

32b Black dotted asperities distributed on anterior andmid region of the gular flap.

53a Gular flap bilobed, species present in Malawi andMozambique.

H. spingularis3b Gular flap not bilobed, species present in Tanzania.

44a Gular flap rounded with posterior and anterior endsmore equal. The gular flap is usually either equal orwider than height, species found in Southern Highlandsof Tanzania.

H. davenporti4b Gular flap narrowly tapering anteriorly and usuallyequal or greater in height, species found in EastUsambara, Nguru, and Uluguru Mountains.

H. burgessi5a Gular flap not bilobed and found in UdzungwaMountains. Females reach a moderate size 18–24 mm.

H. minutissimus5b Gular flap bilobed, and found in Rubeho Mountains.Females reach a large size >25 mm.

H. ukwiva

1

1

Loader et al. BMC Research Notes (2015) 8:167 Page 15 of 16

ConclusionsBased on the description of three new species and refine-ment of the distribution of Hyperolius minutissimus andH. spinigularis some conclusions can be made on the di-versity and distribution of this group of hyperoliids in EastAfrica. The geographic extent of H. spinigularis can nowbe strictly confined to Malawi and Mozambique – in linewith earlier thoughts outlined by Schiøtz [2] who sug-gested that the northern Tanzanian population might rep-resent a distinct lineage. Our molecular results indicatethe Malawi and Mozambique populations of H. spinigu-laris exhibit a sister relationship. However, the morpho-logical similarities between Malawi and MozambiqueH. spinigularis cannot currently be evaluated as only a sin-gle juvenile specimen has been collected from Mt. Namuliin Mozambique [6], and was therefore excluded frommorphometric analyses. An adult has been photographedand resembles the colour pattern of the juvenile, includinga distinctive heel spot (W. Conradie, pers. comm.). Carefulassessment of other characters will be necessary to assesswhether the Mozambique population merits species rec-ognition. The strong biogeographic ties between Mt.Mulanje, Mt. Namuli, and other smaller Mozambicanmassifs have only recently been highlighted by phylogen-etic studies of other endemic montane taxa, includingdwarf day geckos [21] and pygmy chameleons [22]. It ispossible that continued survey work may uncover add-itional populations of H. spinigularis occurring on thesmaller massifs of Mozambique, which can then beassessed in the currently presented phylogenetic and mor-phological framework.Most species in the spiny-throated group exhibit rela-

tively small distributions, with H. tanneri, H. davenporti,and H. ukwiva showing distributions restricted to two orless localities. Current estimates of species range areasare particularly narrow and given the limited remainingforest area in these areas and most of the species associ-ation with forest (apart from H. minutissimus), speciesare likely to be threatened by increasing habitat change.It will be important with future sampling of these areas toassess if this is truly the full extent of their distributions orif they are indeed wider, as these scenarios have importantconservation implications. Future preservation of theseendemic species might require specific species-level con-servation management approaches. However, the strat-egies can also account for the many other threatenedspecies in the area – and provide more assemblage orhabitat-wide level management approaches.

EthicsAll specimens were collected according to internationalguidelines and research permits to collect and exportspecimens from relevant countries were acquired (seealso acknowledgements).

Zoobank numbersHyperolius burgessiurn:lsid:zoobank.org:act:568E4D94-9 F08-417D-81FB-9BF9F230B7BB.

Hyperolius davenportiurn:lsid:zoobank.org:act:61C3C607-1 F44-4CF9-9885-18990DD5337F.

Hyperolius ukwivaurn:lsid:zoobank.org:act:EEF8A317-0C54-44A6-ACA7-93D59259108D.

Additional files

Additional file 1: Supplementary data on species designations,locality information, and photos of phenotype variation withinspecies.

Additional file 2: Morphological measurements for all specimens.

Competing interestsThe authors declare that they have no competing interests.

Authors’ contributionsAll authors contributed, and read and approved the final manuscript.

AcknowledgementsFor advice, help with fieldwork, permits for research and export in Tanzania,we thank (no particular order) Tanzania Commission for Science andTechnology (COSTECH research permit RCA 2001–272; RCA 2007–153, RCA2009-306-NA-2009-201, 2011-239-NA-2011-82, 2006 and 2007-72-Na-2006-19),Tanzania Wildlife Research Institute (TAWIRI), and Wildlife Division for grantingpermission to conduct research in Tanzania and export these specimens. DMPthanks Mozambican officials acknowledged in Portik et al. 2013a for permittingand logistics, which resulted in a research permit administered by theUniversidade Eduardo Mondlane Natural History Museum of Maputo (No. 04/2011 and 05/2011) and a credential administered by the Ministry of Agriculture.Mozambican herpetological specimens were exported to the Museum ofVertebrate Zoology, University of California, Berkeley, under CITES Permit No.MZ-0354/2011. In Malawi, the Museums of Malawi for Malawian assisted inresearch permit HERP/110/267. People thanked in Lawson (2010) are thankedagain for supporting this research. Mark Wilkinson and Patrick Campbell (BMNH)Alan Resetar (FMNH) and Jens Vindum (CAS) are thanked for assisting in loaningof specimens or access to institutional facilities for making measures.We are also grateful to many people and organisations that providedassistance in the field, logistical support and advice, including GabrielaBittencourt-Silva, Werner Conradie, Kim Howell, Wilirk Ngalason, SandraDürrenberger, Sandra Rudolf, Sophy Machaga, Noah Mpunga, Frontier-Tanzania, Tanzania Forest Conservation Group, East Usambara ConservationArea Management Project and the Wildlife Conservation Societies SouthernHighlands Conservation Programme. This work was funded by variousorganisations including, the Swiss National Science Foundation (grantnumber 31003A-133067 to SPL), SCNAT and Freiwillige AkademischeGesellschaft Basel. DMP is supported by an NSF DDIG grant (DEB 1311006).The University of Basel Kick Start Grant University of Chicago, the FieldMuseum of Natural History Africa Council, and Museum of VertebrateZoology also funded surveys and lab work.

Author details1Department of Environmental Sciences, University of Basel, BiogeographyResearch Group, Basel 4056, Switzerland. 2Committee on EvolutionaryBiology, University of Chicago, 1025 E. 57th Street, Culver Hall 402, Chicago,IL 60637, USA. 3Field Museum of Natural History, 1400 S. Lake Shore Drive,Chicago, IL 60605, USA. 4University of Cincinnati, 614 Rieveschl Hall,Cincinnati, OH 45221, USA. 5Museum of Vertebrate Zoology and Department

Loader et al. BMC Research Notes (2015) 8:167 Page 16 of 16

of Integrative Biology, University of California, 3101 Valley Life SciencesBuilding, Berkeley, California 94720, USA. 6Tropical Biodiversity Section,Science Museo of Trento, Via della Scienza e del lavoro, 38122 Trento, Italy.

Received: 10 December 2014 Accepted: 11 March 2015

References1. Lawson LP. The discordance of diversification: evolution in the tropical-

montane frogs of the Eastern Arc Mountains of Tanzania. Mol Ecol.2010;19:4046–60.

2. Schiøtz A. The Treefrogs of Eastern Africa. Copenhagen: Steenstrupia; 1975.3. Stevens RA. A new treefrog from Malawi. Zool Afr. 1971;6:313–20.4. Schiøtz A, Westergaard MM. Notes on some Hyperolius (Anura:

Hyperoliidae) from Tanzania, with supplementary information on tworecently described species. Steenstrupia. 2000;25:1–9.

5. Menegon M, Doggart N, Owen N. The Nguru mountains of Tanzania, anoutstanding hotspot of herpetofaunal diversity. Acta Herpetologica.2008;3:107–27.

6. Portik DM, Mulungu E, Sequeira D, McEntee JP. Herpetological Surveys ofthe Serra Jeci and Namuli Massifs, Mozambique, and an AnnotatedChecklist of the Southern Afromontane Archipelago. Herpetologic Rev.2013;44:394–406.

7. Lawson LP, Bates J, Menegon M, Loader, SP. Divergence at the edges:Peripatric isolation in the montane Spiny Throated Reed Frog complex.BMC Evol Biol. In review. in press 2015.

8. Stamatakis A. RAxML-VI-HPC: maximum likelihood-based phylogeneticanalyses with thousands of taxa and mixed models. Bioinformatics.2006;22:2688–90.

9. Drummond AJ, Suchard MA, Xie D, Rambaut A. Bayesian phylogenetics withBEAUti and the BEAST 1.7. Mol Biol Evol. 2012;29:1969–73.

10. Yang Z, Rannala B. Bayesian species delimitation using multilocus sequencedata. Proc Natl Acad Sci. 2010;107(20):9264–9.

11. Heled J, Drummond AJ. Bayesian inference of species trees from multilocusdata. Mol Biol Evol. 2010;27:570–80.

12. Reid NM, Carstens BC. Phylogenetic estimation error can decrease theaccuracy of species delimitation: a Bayesian implementation of the generalmixed Yule-coalescent model. BMC Evol Biol. 2012;12:196.

13. Grummer JA, Bryson RW, Reeder TW. Species delimitation using Bayesfactors: simulations and application to the Sceloporus scalaris species group(Squamata: Phrynosomatidae). Syst Biol. 2013;63(2):119–33.

14. Team RC. R: A language and environment for statistical computing. 2012.15. Wickham H. Ggplot2: Elegant Graphics for Data Analysis. New York: Springer; 2009.16. Venables WN, Ripley BD. Modern Applied Statistics with S. New York:

Springer; 2002.17. Charif RA, Clark CW, Fristrup KM. Raven 1.2 user’s manual. Ithaca, New York:

Cornell Laboratory of Ornithology; 2004.18. Vonesh JR. Sequential predator effects across three life stages of the African

tree frog, Hyperolius spinigularis. Oecologia. 2005;143:280–90.19. Rovero F, Menegon M, FjeldsAa J, Collett L, Doggart L, Leonard C, et al.

Targeted vertebrate surveys enhance the faunal importance and improveexplanatory models within the Eastern Arc Mountains of Kenya andTanzania. Divers Distrib. 2014;20(12):1438–49.

20. Schiøtz A: Treefrogs of Africa. Frankfurt/Main: Chimaira; 1999. 350 pp.21. Portik DM, Travers SL, Bauer AM, Branch WR. A new species of Lygodactylus

(Squamata: Gekkonidae) endemic to Mount Namuli, an isolated “sky island”of northern Mozambique. Zootaxa. 2013;3710:415–35.

22. Branch WR, Bayliss J, Tolley KA. Pygmy chameleons of the Rhampholeonplatyceps compex (Squamata: Chamaeleonidae): Description of four newspecies from isolated “sky islands” of northern Mozambique. Zootaxa.2014;3814:1–36.

Submit your next manuscript to BioMed Centraland take full advantage of:

• Convenient online submission

• Thorough peer review

• No space constraints or color figure charges

• Immediate publication on acceptance

• Inclusion in PubMed, CAS, Scopus and Google Scholar

• Research which is freely available for redistribution

Submit your manuscript at www.biomedcentral.com/submit